Cement & Concrete Composites 22 (2000) 453±458

www.elsevier.com/locate/cemconcomp

E€ect of hot-dry curing environment on the intrinsic properties of

repair materials
K.E. Hassan a,*, P.C. Robery a,b, L. Al-Alawi a
a
Civil Engineering Materials Unit (CEMU), School of Civil Engineering, University of Leeds, Leeds LS2 9JT, UK
b
Maunsell Ltd, Attwood House, 1 Worcester Walk, Birmingham, B2 4NX, UK
Received 10 February 2000; accepted 29 August 2000

Abstract
The paper examines the properties of ®ve di€erent types of repair materials, including conventional cementitious, polymer and
polymer-modi®ed repair mortars. Assessment was carried out on the basis of the engineering properties (compressive strength,
tensile strength and modulus of elasticity), pore structure (porosity and pore size distribution), transport properties (permeability
and di€usion) and shrinkage. These properties were measured up to the age of 28 days after curing in a hot-dry environment.
The epoxy resin repair mortar showed superior strength and transport characteristics with a very ®ne pore structure; however, its
modulus of elasticity was remarkably low when compared with that of normal- and high-strength concretes. A hot-dry curing
environment adversely a€ects the shrinkage and performance-related properties of conventional repair mortars; however, small
improvements could be achieved by the use of mineral admixtures (¯y ash and silica fume). The paper discusses also the di€erent
testing techniques which could be used to assess the potential performance of concrete repair mortars. Ó 2000 Elsevier Science Ltd.
All rights reserved.

Keywords: Concrete; Curing; Elastic modulus of elasticity; Di€usion; Hot-dry environment; Mineral admixtures; Permeability; Repair mortars;
Shrinkage; Strength

1. Introduction level of shrinkage and cracking can be expected, espe-

Corrosion of reinforcements is still a major cause of
deterioration in concrete structures. The high cost in-
curred in repairing and strengthening faulty structures
makes it essential that the materials used will provide the
durability to match the residual design life of the
structure. This is, of course, a strong reason for inves-
tigating and developing improved repair materials and
testing methodologies.
Depending on the type of exposure environment and
the condition of the deteriorated structure, selection of
the appropriate repair mortar can be based on its in-
trinsic properties as well as its compatibility with the
existing structure. Conventional repair mortars made
using cements and sand will be similar to the existing
concrete structure in terms of engineering and perfor-
mance properties, cost and appearance. However, a high
*
Corresponding author. Fax: +44-113-233-2265.
E-mail address: k.e.hassan@leeds.ac.uk (K.E. Hassan).
cially when cured in a hot-dry environment [1,2].
Manufacturers prepare specially formulated
``shrinkage-compensating'' mortars, designed to com-
bat these problems (polymer-modi®ed mortars, PMC).
These mortars often contain polymer in the range of
5±10%, additives and polypropylene ®bres to yield
improved properties and reduced cracking [3]. They
are considered as very e€ective repair materials [4]
due to their improved performance properties (high
resistance to the penetration of chloride ions, oxygen
and carbon dioxide). However, their shrinkage defor-
mation is signi®cantly a€ected by the curing environ-
ment [5].
Only polymer mortars o€er real independence from
elevated temperatures, as their resin base is immune to
moisture loss. These products can be formulated to gel
and cure rapidly over a wide temperature range. They
provide excellent strength and bond to concrete, with
low shrinkage and permeability [2,3]. However, their
structural and thermal compatibility are signi®cantly
di€erent from those of conventional concrete [6,7].

0958-9465/00/$ - see front matter Ó 2000 Elsevier Science Ltd. All rights reserved.
PII: S 0 9 5 8 - 9 4 6 5 ( 0 0 ) 0 0 0 4 8 - 2
454 K.E. Hassan et al. / Cement & Concrete Composites 22 (2000) 453±458

This paper focuses on the intrinsic properties of ®ve
common types of repair materials, whereas the com-
patibility of these materials with concrete is investigated
in another study [8]. Measurements of pore structure are
obtained and used for the interpretation of the engi-
neering, transport and shrinkage properties under a hot-
dry curing environment.
2. Repair materials
Five repair materials were selected in this study. These
include conventional cementitious, epoxy resin (EP) and
PMC. Table 1 gives details of the repair materials.
The conventional repair mortars were proportioned
on the basis of minimum porosity by achieving maxi-
mum packing of the binder and sand particles. A dosage
of 3% (weight of binder) of a melamine formaldehyde-
based superplasticiser was used, and the amount of
mixing water was adjusted to have similar workability
values as those obtained by the ¯ow-table test according
to BS 4550. Table 1 shows also the workability values
for the di€erent repair mortars. The commercial pre-
packed PMC and EP repair mortars were mixed and
cast according to the manufacturer's instructions.
3. Testing programme
After mixing the repair materials, they were cast,
compacted in their moulds on a vibrating table and left
overnight covered with wet hessian and polyethylene
sheets. On the following day they were de-moulded and
transferred to a hot-dry environmental chamber main-
tained at 35°C, 45% relative humidity and 3 m/s wind
velocity. These conditions were selected to examine the
e€ects of extreme curing on the properties of repair
materials.
Compressive strength, modulus of elasticity and ten-
sile strength testing were carried out to study the engi-
neering properties of the di€erent repair mortars. Cubes
(50 mm sides) were used for the measurement of com-
pressive strength at 1, 3, 7 and 28 days. The static
compressive modulus of elasticity test was performed on
100 mm3 cubes at 28 days. Strain gauges were ®tted on
the side of the cube specimens, and the modulus was
calculated for each material from the stress±strain rela-
tionship up to 1/3 of the failure load. Previous work [9]
indicated that specimen size has no in¯uence on the
sti€ness and the stress±strain curve. The direct tensile
strength was measured using bobbin-shaped specimens
(75 mm diameter and 325 mm height) at 28 days [10],
using the Instron 8500 Series Digital testing instrument.
Mortar slabs (400  250  50 mm3 ) were prepared
for the measurement of transport properties and pore
size distribution at similar testing ages to those of the
compressive strength. Porosity, oxygen permeability,
water permeability and oxygen di€usion testing were
conducted on 50 mm diameter cores, whereas 20 mm
diameter cores were used for the measurement of pore
size distribution using the technique of mercury intru-
sion porosimetry. Porosity testing was carried out using
the method of vacuum-saturation with water [11]. For
the oxygen permeability testing, the specimens were
prepared and tested as described in Ref. [12]. The same
specimens were also used for the measurement of oxygen
di€usion and water permeability. Details of the perme-
ability and di€usion cells with the testing procedures are
described in Refs. [11±13].
Unrestrained shrinkage of the repair mortars was
carried out using cylindrical specimens (75 mm diameter
and 265 mm height) from two days after casting the
specimens up to the age of 28 days. Demec points were
attached to the curved surface of each cylinder across a
length of 200 mm as described in ACI 209R-92 [14].
4. Results and discussion
4.1. Mechanical properties
4.1.1. Compressive strength
The compressive strength results of the di€erent re-
pair mortars are presented in Fig. 1. In general, EP

Table 1
Repair materials used in the study

Code Description Flow-table spread (mm)

OPC Ordinary Portland cement: OPC and sand in the weight ratio of 1:2.33. Water/cement of 0.33. 210
Superplasticiser: naphthalene sulphonated polymer-based admixture (3 litres per 100 kg of binder).
FA Fly Ash mortar: similar to the OPC mortar, replacing 30% of OPC with FA. Water/binder of 0.30. 200
SF Silica Fume mortar: similar to the OPC mortar, replacing 10% of OPC with SF. Water/binder of 0.33. 200
PMCa Polymer-modi®ed mortar: a commercial two-component (A and B) ®bre-reinforced polymer-modi®ed 190
mortar. Component A: acrylic copolymer; component B: blend of cements/aggregates/admixtures.
EPa Epoxy resin mortar: a commercial three-component epoxy resin-based repair mortar (A: resin, B: 110
hardener and C: aggregate).
a
Supplied by the manufacturer.
 K.E. Hassan et al. / Cement & Concrete Composites 22 (2000) 453±458 455

Fig. 2. 28-day tensile strength of the di€erent repair mortars.

Fig. 1. Compressive strength of the di€erent repair mortars.
(polymer repair mortar) showed the highest compressive
strength at all ages, whereas the conventional repair
mortars (OPC, FA and SF) gave the lowest values. PMC
(polymer-modi®ed repair mortar) showed values inter-
mediate between the polymer and the conventional re-
pair materials.
The e€ect of curing environment on the 28-day
compressive strength is presented in Table 2, which gives
also the compressive strength of the conventional mor-
tars when cured at 20°C and 99% relative humidity [15].
By comparing the results, it can be seen that the curing
environment has only little e€ect on the strength of both
PMC and EP mortars. The case was di€erent for the
conventional repair mortars, where the hot-dry envi-
ronment (35°C and 45% RH) has adversely a€ected the
strength. A reduction of about 50% of the strength oc-
curs for the same mortars when cured under a hot-dry
environment.
ratio (1/10) to its compressive strength. A superior
strength was found for the polymer repair mortar (EP),
which is only 1/5 of its compressive strength value and
about four times higher than that of the conventional
mortars.
4.1.3. Modulus of elasticity
The modulus of elasticity is an important property
regarding the deformability of repair materials under
load conditions. The compressive static moduli of elas-
ticity of the di€erent repair materials are graphically il-
lustrated in Fig. 3, and also given in Table 2. In contrast
to the strength results, the EP repair mortar showed the
lowest modulus value of 13 GPa. The highest value was
obtained for PMC (41 GPa), while the conventional
repair mortars (OPC, FA and SF) showed almost sim-
ilar values between 29 and 32 GPa.
4.2. Porosity and pore size distribution

The results of total porosity, as obtained from the

4.1.2. Tensile strength vacuum saturation test, are illustrated in Fig. 4. The
Fig. 2 and Table 2 present the 28-day results of the porosity values varied widely (0.5±15%) for the di€erent
direct tensile strength for the di€erent repair materials. types of repair mortars used in this study. The polymer
The conventional mortars showed a tensile strength repair mortar (EP), which showed the highest strength
value of about 4 MPa, which is approximately 1/10 of properties, exhibited the lowest porosity values. In fact,
their compressive strength values. PMC showed im- the porosity of EP was extremely low (<1%) at all tested
proved tensile strength (6 MPa), with almost the same ages. The highest porosity values were obtained for the
conventional repair mortars, which exhibited higher
Table 2
Strength and modulus properties at 28 days
porosity values than the PMC. Among the conventional
mortars, the SF gave the lowest compressive strength
Compressive Tensile ft =fc Modulus and also the lowest porosity.
fc (MPa) ft (MPa) (%) E (GPa) Measurements of pore size distribution combined
Curing 35°C 20°C 35°C 35°C 35°C
with total porosity are essential for the pore structure
OPC 42.8 (83.6) 4.08 9.5 32.3 analysis of repair mortars. There are many parameters
FA 47.6 (88.6) 3.84 8.1 28.6 that can be used to represent the pore size distribution
SF 39.0 (90.7) 3.48 8.9 31.4 results. In this study, the median pore diameter (MPD),
PMC 65.4 [>55.0] 5.94 [7.6] 9.1 41.4 which corresponds to the pore size at 50% intrusion, was
EP 88.8 [>80.0] 17.51 [18] 19.7 13.2
used. Fig. 5 shows the MPD for the di€erent repair
( ): at 20°C and 99% relative humidity, Ref. [15]; [ ]: at 20°C and 65% materials up to 28 days. The polymer repair mortar
relative humidity, supplied by the manufacturers. exhibited not only the lowest porosity value but also the
456 K.E. Hassan et al. / Cement & Concrete Composites 22 (2000) 453±458

Fig. 3. 28-day compressive modulus of elasticity.

Fig. 5. Median pore diameter of the repair mortars.

Fig. 4. Porosity vs. age for the di€erent repair mortars.

Fig. 6. Variation of oxygen permeability for the di€erent repair mor-

®nest pore structure, which is at least 10 times smaller tars.
than that of conventional repair mortars. For PMC, the
MPD reduced sharply within the ®rst week, indicating
the re®nement of pore sizes even under hot-dry envi-
ronmental conditions. The e€ect of pore structure on the
performance properties of the repair materials is dis-
cussed in the following sections.

4.3. Transport properties

4.3.1. Permeability
Permeability is a property which de®nes the resistance
to the penetration of aggressive substances under the
in¯uence of di€erential pressure and is fundamental
regarding concrete durability and long-term perfor-
mance. The values of oxygen and water permeability are
shown in Figs. 6 and 7, respectively.
In general, the results varied widely (3±4 orders of Fig. 7. Water permeability vs. age for the di€erent repair mortars.
magnitude) for the di€erent repair materials. The EP
repair mortar, with its ®ne pore structure, showed the
lowest permeability values. These values are extremely 4.3.2. Oxygen di€usion
low when compared to high-performance concrete [16]. Di€usion is another transport property describing the
The permeability of PMC was about one order of rate of movement of aggressive substances due to a
magnitude lower than that of the conventional repair concentration gradient. The oxygen di€usion values
mortars. measured for the di€erent materials are presented in
 K.E. Hassan et al. / Cement & Concrete Composites 22 (2000) 453±458
Fig. 8. Oxygen di€usion of the di€erent repair mortars.
Fig. 9. Shrinkage of the di€erent repair mortars.
Fig. 8. The polymer repair mortar (EP) showed the
lowest di€usion value whereas the conventional repair
mortars exhibited the highest values, similar to the
permeability results.
4.4. Shrinkage
Shrinkage is a volume-change property which is
measured by monitoring periodically the length changes
of the specimens exposed to controlled environmental
conditions. The specimen size and the exposure condi-
tions, which vary in di€erent standards, have a great
in¯uence on the shrinkage of repair mortars [17]. The
drying shrinkage of the di€erent repair mortars is shown
in Fig. 9. The results show clearly that conventional
repair mortars are more sensitive to the hot-dry envi-
ronment, which accelerates the moisture loss and en-
hances shrinkage. Dramatic reduction in shrinkage can
be obtained using polymers. PMC showed a reduction
of more than 50% compared to the conventional mor-
tars. Further reduction can be obtained using polymer
mortar (EP). The latter showed less than 20% shrinkage
compared to the conventional mortars.
457
5. Discussion
In this study, the intrinsic properties measured for the
repair mortars varied widely depending on their generic
types and formulations, which result in di€erent pore
structure in¯uencing both engineering and performance
properties of repair mortars. The epoxy resin (polymer)
mortar, exhibiting the lowest porosity values and the
®nest pore sizes, showed further improved characteristics.
The coarse pore structure of the cementitious mortars
resulted in lower compressive and tensile strength values,
which were about two and four times lower than those of
the epoxy mortar, respectively. However, the di€erence
was further enhanced in the transport properties, where
the reduction was about three orders of magnitude in the
permeability and two orders of magnitude in the di€u-
sion. This indicates clearly that the transport properties
are more representative of mortar pore structure and
strongly supports the adoption of such properties within
the performance-based criteria of repair mortars. Cab-
rera and Al-Hassan [18] reported that measurements of
permeability and pore structure are the most rational way
of assessing the quality of repair mortars.
The drying shrinkage data obtained in this study in-
dicated the sensitivity of the conventional mortars to
curing environment. The fast withdrawal of moisture
due to a hot-dry environment adversely a€ects the dry-
ing shrinkage to give values about 6±8 times and 2±3
times higher than those of polymer and polymer-modi-
®ed repair mortars, respectively. However, the use of
mineral admixtures (FA and SF) was found to improve
the pore structure and consequently reduce the shrink-
age. A di€erent deformation behaviour is obtained for
the repair mortars under load application. Mangat and
O'Flaherty [19] indicated that, for the design of ecient
repair, the repair material should have greater modulus
than the concrete substrate. The modulus of elasticity of
the conventional repair mortars was quite similar to that
of normal concrete. The epoxy mortar gave the lowest
value, about 1/3 that of normal concrete, indicating
much more deformation under load application.
Currently there are developments towards uni®ed
European Standards for repair materials. These stan-
dards, as reviewed by Robery and Shaw [20,21], are not
limited to the strength properties but extended to dif-
ferent characteristics such as: composition, usability and
physical (thermal compatibility, dimensional stability
and pull-o€ adhesion) and transport (resistance to
chloride, carbonation and water penetration) properties.
Most of these performance tests are stated to be con-
ducted after curing the specimens at 21  2°C. However,
and based on the results obtained in this study, great
care should be taken with the use of repair materials,
especially cementitious mortars, when exposed to hot-
dry environmental conditions.
458 K.E. Hassan et al. / Cement & Concrete Composites 22 (2000) 453±458
6. Conclusions
1. The intrinsic properties of the repair mortars vary
widely depending on their formulations and the expo-
sure environment.
2. Under a hot-dry curing environment, polymer and
polymer-modi®ed repair materials exhibit improved
strength, permeability, di€usion and shrinkage prop-
erties when compared to conventional cementitious
repair mortars.
3. Di€erences between the di€erent types of repair mate-
rials studied are more pronounced in their transport
properties than their strength properties, indicating
the importance of the permeability and di€usion
characteristics in the performance speci®cations of re-
pair mortars.
4. Within the di€erent types of repair mortars used in
this study, the polymer (resin) material deforms more
under load application, indicating incompatibility
with normal- and high-strength/performance con-
cretes.
References
[1] Tabor LJ. Repair materials and techniques. In: Mays G, editor.
Durability of concrete structures: investigation, repair, protec-
tion. E&FN Spon; 1992. p. 82±129.
[2] Perkins PH. Repair, protection and waterproo®ng of concrete
structures. London: Elsevier Applied Science; 1986.
[3] Fowler DW. Polymer in concrete: a vision for the 21st century.
Cem Concr Compos 1999;21(5±6):449±52.
[4] Ohama Y. Polymer-based materials for repair and improved
durability: Japanese experience. Constr Build Mater
1996;10(1):77±82.
[5] Mangat PS, Limbachiya MC. Repair material properties for
e€ective structural application. Cem Concr Res 1997;27(4):601±
17.
[6] Abu-Tair AL, Burley E, Rigden SR. Heat cycling of epoxy
mortar repaired beams. ACI Mater J 1997;94(2):129±33.
[7] Saccani A, Magnaghi V. Durability of epoxy resin-based mate-
rials for the repair of damaged cementitous composites. Cem
Concr Res 1999;29(1):95±8.
[8] Hassan KE, Brooks JJ, Al-Alawi L. Compatibility of repair
materials and concrete in a hot dry environment. School of Civil
Engineering Report, University of Leeds, 2000, 24 pp.
[9] Chin MS, Mansur MA, Wee TH. E€ects of shape, size, and
casting direction of specimens on stress±strain curves of high-
strength concrete. ACI Mater J 1997;94(3):209±19.
[10] Brooks JJ, Neville AM. A comparison of creep, elasticity and
strength of concrete in tension and in compression. Mag Concr
Res 1977;29(100):131±41.
[11] Cabrera JG. Design and production of high performance durable
concrete. In: Swamy RN, editor. Proceedings of International
Conference. Sheeld: Sheeld Academic Press; 1999. p. 1±14.
[12] Cabrera JG, Lynsdale CJ. A new permeameter for measuring the
permeability of mortar and concrete. Mag Concr Res
1988;40(114):177±82.
[13] Cabrera JG and Hassan KE. Controlling the quality of concrete
by measuring its permeability. In: Proceedings of the 13th
International Conference on Building Materials, Weimar, 1997,
vol. 3. p. 3/005±3/017.
[14] ACI Committee 209R-92. Prediction of creep, shrinkage, and
temperature e€ects in concrete structures. ACI (USA); 1996.
[15] Hassan KE, Cabrera JG, Ahmed SA. E€ectiveness of repair
materials for deteriorated concrete structures. In: First Interna-
tional Civil Engineering Egypt±China±Canada Symposium on
Evaluation, Repair and Retro®t of Structures using Advanced
Methods and Materials, Cairo, 1997. p. 37±42.
[16] Hassan KE, Cabrera JG, Maliehe RS. The e€ect of mineral
admixtures on the properties of high performance concrete. Cem
Concr Compos 2000;22(4):267±71.
[17] Decter MH, Keeley C. Durable concrete repair ± importance of
compatibility and low shrinkage. Constr Build Mater 1997;11(5±
6):267±73.
[18] Cabrera JG, Al-Hassan AS. Performance properties of concrete
repair materials. Constr Build Mater 1997;11(5±6):283±90.
[19] Mangat PS, O'Flaherty FJ. In¯uence of elastic modulus on stress
redistribution and cracking in repair patches. Cem Concr Res
2000;30(1):125±36.
[20] Robery PC. Standards for concrete repair and protection. In:
Kruger D, editor, Proc. Int. Workshop on Concrete Repair,
Johannesburg, 2000, pp. 87±104.
[21] Robery PC, Shaw JDN. Materials for the repair and protection of
concrete. Constr Build Mater 1997;11(5±6):275±81.